of the royal food on the inner margin of the worker cell and produced queens. With bees, then, we see that the neuter is only a female with the reproductive organs partially developed, a conclusion further enforced by the fact that the neuter sometimes even lays eggs which develop into the drone or male bee. In this case then, at least, it may be said that the fecundity or sterility of bees depends almost entirely on the nature of the food given to the larvæ. With the true ants also, the so-called neuter is only a partially developed female, although there are no recorded observations to show that the difference between the fertile and the infertile insect is due to difference in food.
To advance now a step further: insects, originally, were either male or female; there were no neuters. For reasons presently to be given, we conclude that the males of even social insects, in addition to the duty of fecundating the female, also undertook the duty of defending the nest against warlike intruders. It may be also safely concluded that originally the females of even social insects, in addition to the duties of maternity, also undertook the labors of building the nest and taking care of the young—customs that still prevail more or less with many insects. Thus, the females of the solitary bees, after impregnation, hybernate during the winter. With the warm days of spring they awaken, build their nests, and die. Among wasps a similar state of things obtains. When the winter approaches, the entire colony, with the exception of a few pregnant females, dies. In the spring these females begin building a new nest, laying eggs, and so producing an efficient corps of assistants to aid them in their future labors.
Supposing, then, that originally the females undertook most of the work now done by the neuters; supposing, also, that at the beginning of spring there are only a few females, or even one female, to commence the work of building the nest and feeding the future progeny. With many duties to perform, it is not unwarranted to conclude that some duties may not be completely performed. The immediate success of the colony depends, not so much on the number of males or of females, as on a body of efficient assistants. Now, when one or a few insects have to feed many, some of the larvæ receive an abundance, some barely a sufficiency of food; and, on the theory previously advanced, we may see how neuter insects arose. The fact that quality as well as quantity of food is essential to future fertility does not add to the difficulty, since, as it takes more time to produce the highly nutritious than the somewhat less nutritious food, the conditions are still the same. If efficient assistants could be produced at a less expenditure of labor and in a shorter time than females, a supposition countenanced by facts, natural selection begins to work. In a given neighborhood those insects which produce a corps of assistants soonest, and with least expenditure of labor, will stand a better chance of obtaining food—of surviving—than those insects which give the same amount
